Ischemia-induced cleavage of cadherins in NRK cells requires MT1-MMP (MMP-14)

Marisa D. Covington, Robert C. Burghardt, Alan R. Parrish

Abstract

Ischemia is a leading cause of acute renal failure (ARF), a disease associated with high morbidity and mortality. Disruption of intercellular adhesion in the proximal tubules is linked to ARF, although the molecular mechanism(s) remains unclear. Our previous studies showed that ischemia is associated with cadherin cleavage and loss in NRK cells, putatively due to a matrix metalloproteinase (MMP) (7). In the current studies, a MMP required for E-cadherin cleavage and N-cadherin loss was identified. Chemical inhibitors against a number of soluble MMPs (1, 2, 3, 8, 9) failed to completely attenuate ischemia-induced cadherin loss. Under ischemic conditions, there was an increase in active membrane-type (MT)1-MMP but a decrease in MMP-2 protein expression. Plating cells on fibronectin protected against ischemia-induced loss of cadherins and, interestingly, no increase in active MT1-MMP levels was seen in ischemic cells on fibronectin-coated dishes. In addition, L cells stably expressing E- (LE) or N-cadherin (LN), but lacking MT1-MMP expression, were resistant to ischemia-induced cadherin loss. The role of MT1-MMP in ischemia-induced cadherin loss was confirmed by either blocking MT1-MMP activity with a neutralizing antibody or expression with shRNA constructs which protected full-length E- and N-cadherin during ischemia. Using shRNA constructs to suppress MT1-MMP expression, ischemia-induced disruption of cadherin function was ablated, and cell-cell contacts were preserved. These results demonstrate that ischemia induces increased expression of active MT1-MMP and subsequent disruption of cadherin/catenin complexes, implying that MT1-MMP plays a role in ischemia-induced ARF.

  • shRNA
  • acute renal failure
  • proximal tubules

during the past 50 years, the morbidity due to acute renal failure (ARF) has not improved, and the mortality rate can be as high as 70% (46). In addition, the development of ARF increases the mortality associated with other diseases. As disruption of cadherin/catenin complexes can account for several of the tubular changes including loss of polarity and cytoskeletal disorganization which are critical to ARF (reviewed in Ref. 3), we have focused on elucidating the role of cadherin/catenin complexes in renal pathophysiology (7, 16, 17).

Molitoris and Marrs (29) reviewed the evidence for a role of cell adhesion molecules, including cadherins, in ARF. The finding that localization of cadherins and catenins is altered in patients with sustained ARF associated with transtubular backleak of glomerular filtrate strengthens the hypothesis that disruption of cadherin/catenin complexes is associated with ARF (24). Mechanistically, internalization of E-cadherin was seen in Madin-Darby canine kidney (MDCK) cells challenged with antimycin A and 2-deoxyglucose to deplete ATP (25). Bush et al. (4) showed that chemical ischemia (antimycin A/2-deoxyglucose) caused a loss of cell-surface E-cadherin and degradation of E-cadherin to a fragment of ∼80 kDa, which was not blocked by inhibitors of the proteasomal, lysosomal, or calpain proteolytic pathways (4). Our laboratory showed that ischemia is associated with activation of a matrix metalloproteinase (MMP) that disrupts cadherin expression and function in the NRK-52E cell line (7).

MMPs are Zn2+- and Ca2+-dependent enzymes that are synthesized as zymogens; once activated, they regulate extracellular matrix (ECM) degradation (33, 34). Traditionally, MMP substrates were thought to be limited to ECM components, but this view has changed with the discovery of non-ECM substrates. Cadherins are cleaved by MMPs as first demonstrated by Bissell and colleagues (24) who showed that induction of stromelysin-1 (MMP-3) in mammary epithelial cells was associated with cleavage of E-cadherin to an 80-kDa extracellular fragment. Since that time, a number of investigators have shown that MMP-3 (36, 39) and MMP-7 (27, 36) can cleave E-cadherin. VE- (55) and N-cadherin (32, 53) have also been identified as MMP substrates, suggesting that cadherins represent an important class of proteins regulated by MMPs.

The membrane-type MMPs (MT-MMPs) are a family of membrane-anchored MMPs consisting of six members: MT(1–6)-MMP (reviewed in Ref. 13). Structurally, they are similar to the MMPs with the exception that they either contain a transmembrane domain and short cytoplasmic tail (MT1-, MT2-, MT3-, and MT5-MMP), or they are anchored to the membrane via a glycosylphosphatidyl inositol anchor (MT4- and MT6-MMP). In addition to cleaving ECM substrates, MT1-MMP regulates the activity of MMP-2 in conjunction with tissue inhibitor of metalloproteinse-1 (TIMP-1) (5, 49), MMP-9 (52), and MMP-13 (19). As opposed to the soluble MMPs, MT-MMPs exhibit significant differences in inhibition by tissue inhibitors of metalloproteinases (TIMPs). MT1- and MT3-MMP are not inhibited by TIMP-1, while MT2-, MT4- and MT6-MMP are inhibited by TIMP-1 (10, 43, 54).

MMPs are thought to be involved in ischemia-induced damage in the brain, lung, and heart (31, 44, 47, 56); however, the role MMPs play in renal injury remains uncertain. In the kidney, MT1-MMP, MT2-MMP, and MT5-MMP are expressed (18, 22, 23, 37), as well as MMP-2, -9, and TIMP-1, -2, -3 (51). Although Ziswiler et al. (57) did not detect an increase in MMPs using zymography in rats after 1 h of ischemia and either 12 or 24 h of reperfusion, Basile et al. (2) demonstrated that MMP-2 and -9 were enhanced at 1–3 days of reperfusion following 52 min of ischemia in rats. MMP-7 is not expressed in the normal adult human kidney but is induced under pathophysiological conditions (50), suggesting that MMPs may be important in renal injury. The objective of the current study was to identify the specific MMP associated with loss of cadherin expression and function in NRK cells following in vitro ischemia.

MATERIALS AND METHODS

Cell culture.

NRK-52E and L-cells, a fibroblast cell line lacking cadherin expression (NCTC clone 929), were purchased from ATCC (Manassas, VA) and cultured on Falcon polystyrene culture dishes in DMEM containing 1.5 g/l sodium bicarbonate and 5% bovine serum in an atmosphere of 5% CO2-95% air at 37°C. At confluence, subcultures were prepared by treatment with 0.02% EDTA, 0.05% trypsin solution, and cells were seeded at a density of 4 × 104 cells/cm2. Cells were used between passages 3 and 16. LE cells (L cells stably expressing E-cadherin) were provided by Dr. J. Balsamo (University of Iowa) and grown in DMEM containing 1.5 g/l sodium bicarbonate, 10% bovine serum, and 100 μg/ml of G418 in an atmosphere of 5% CO2-95% air at 37°C. In certain experiments, NRK cells were grown on precoated dishes (i.e., collagen I, collagen IV, fibronectin, laminin; Biocoat, Bedford, MA) before ischemic insult.

PCR-based cloning of N-cadherin.

Since the sequence of rat N-cadherin is known (6), we used a PCR-based strategy to clone N-cadherin from rat kidneys. Rat kidney tissue was homogenized in Tri-Reagent, chloroform was added, and the sample was vortexed. After centrifugation (12,000 g for 15 min at 4°C), isopropanol was added to the aqueous layer and the sample was placed in a −20°C freezer overnight. RNA was precipitated and a Poly(A)Purist kit (Ambion, Austin, TX) was used to isolate the mRNA. Using RT-PCR, N-cadherin was cloned and inserted into the pCI-neo Vector (Promega, Madison, WI) pCI-neo-N. pCI-neo-N was transformed into JM109 high-efficiency cells (Promega), and isolated colonies were selected and propagated in LB-ampicillin broth. Using maxi prep plasmid purification kits (Qiagen, Valencia, CA), pCI-neo-N was purified and sequenced using a T7 EEV (EukaryoticExpression Vector) promoter primer (Promega) at the Texas A&M Gene Technologies Laboratory. pCI-neo-N was transfected in L-cells using Lipofectamine (Invitrogen, Carlsbad, CA), and stable cell lines were selected using G418. Cells were lysed and a Western blot was performed to confirm N-cadherin expression. LN cells (L cells with stable N-cadherin expression) were maintained in DMEM containing 1.5 g/l sodium bicarbonate, 10% bovine serum, and 100 μg/ml of G418 in an atmosphere of 5% CO2-95% air at 37°C.

Simulated ischemia.

For the current studies, we used a model of in vitro ischemia developed by Meldrum et al. (28) and modified by our laboratory (7). The mineral oil overlay models has the advantages of including both nutrient/oxygen deprivation and metabolite accumulation, which are critical components of ischemia. Confluent cells were washed twice with PBS before the addition of PBS supplemented with 1.5 mM CaCl2 and 2 mM MgCl2. A layer of mineral oil (Sigma 400–5, St. Louis, MO) was added to the culture dish. For a 10-cm2 dish, 2 ml of PBS with Ca2+ and Mg2+ and 10 ml of mineral oil were used. Unless otherwise noted, control cells were incubated for 6 h in PBS with Ca2+ and Mg2+.

Western blots.

Cell culture plates were washed twice with PBS, scraped, and centrifuged. The supernatant was removed, and the cell pellet was washed with cold PBS and then lysed in buffer (10 mM Tris·HCl, pH 7.6, 1% SDS, 1 mM PMSF, 1 mM leuptin, 1 mM orthovanadate) and boiled for 10 min. The homogenates were spun at 18,000 g for 10 min, and the supernatant was collected. Proteins were quantified by the Bradford method and diluted to 1 μg/μl in 2× sample buffer (250 mM Tris·HCl, pH 6.8, 4% SDS, 10% glycerol, 2% β-mercaptoethanol, 0.006% bromophenol blue). For measurement of soluble MMP-2, conditioned PBS was isolated and diluted 1:1 with 2× sample buffer. Samples were boiled for 5 min before electrophoresis, and 20 μg of protein were separated by 8, 12, or 15% SDS-PAGE. Separated proteins were transferred onto a Hybond-ECL nitrocellulose membrane (Amersham, Piscataway, NJ) in transfer buffer (25 mM Tris, 200 mM glycine, 20% methanol, and 1% SDS). Nonspecific binding was blocked by incubation with Tris-buffered saline plus Tween 20 (TBST) blocking buffer (0.1% Tween 20, 10 mM Tris, pH 7.5, 100 mM NaCl) supplemented with 5% nonfat dry milk for 1 h at room temperature. Primary antibodies [E-cadherin (clone 36), N-cadherin (clone 32), β-catenin (clone 14), γ-catenin (clone 15), and p120-catenin (clone 98) from BD Biosciences; α-catenin (C-19) from Santa Cruz Biotechnology; MMP-2 (MAB3308) and MT1-MMP (AB8102) antibodies from Chemicon] were diluted in the same buffer and incubated at 4°C overnight. After subsequent washes with TBST, membranes were incubated with secondary antibody (1:20,000 in TBST:5% nonfat dry milk) against the appropriate species for 1 h at room temperature. The blots were washed three times in TBST and proteins were detected with the Amersham ECL system and exposed to X-ray film (Kodak, Rochester, NY).

Immunofluorescence.

Cells grown on Lab Tek Chamber Slides (Nunc, Rochester, NY) were washed twice with PBS, fixed in 2% paraformaldehyde for 10 min, washed two times in PBS for 10 min, and permeablized in 1% Triton X-100 in PBS for 10 min. Slides were treated with 1:20 blocking solutions of serum related to species in which the secondary antibody was generated at room temperature for 1 h. Primary antibodies were added at appropriate dilutions overnight. After being washed (0.3% Tween in PBS; PBST), FITC-conjugated secondary antibodies (1:200) were added and sections were incubated in the dark at room temperature for 1 h. Slides were mounted with DAPI-O antifade media (Molecular Probes, Eugene, OR) following several washes. Immunostained slides were visualized with a Zeiss Axioplan 2 microscope (Carl Zeiss, Thornwood, NY) fitted with an Axiocam HR digital camera and Axiovision 3.0 software. Negative controls involved substituting IgG for primary antibodies and appropriate species serum for secondary antibodies.

Cell aggregation assay.

The aggregation assay developed for cadherins was utilized where Ca2+ is present during trypsin treatment to preserve cadherin function (11). Normal and ischemic NRK cells were washed with PBS containing 1.5 mM CaCl2 and 2 mM MgCl2, before adding 0.025% trypsin with Ca2+ and Mg2+ for 10 min at 37°C in the presence of 5% CO2. Cells were collected and resuspended at 1.0 × 105 cells/ml in either PBS with 1 mM EDTA or PBS with Ca2+ and Mg2+. From this suspension, 1.5 ml were placed into a 12.5-mm-diameter well coated with 1.5% agarose in PBS. Cells were incubated in a rotary shaker at 80 rpm and 37°C for 20 and 60 min. Aggregated cells were counted manually in a hemacytometer.

Inhibitors.

All inhibitors were added to PBS with Ca2+ and Mg2+ 30 min before ischemia. MT1-MMP blocking antibody was used to inhibit activity of MT1-MMP (2.12 μg/ml, AB8102, Chemicon). MMP inhibitor I, an inhibitor of MMP-1 and -8 (1 μM), -9 (30 μM), and -3 (150 μM), was used at 150 mM; MMP inhibitor III, an inhibitor of MMP-1 (4 nM), MMP-2 (2.3 nM), MMP-13 (10 nM), MMP-7 (10 nM), and MMP-3 (135 nM), was used at 10 nM; and MMP-2 inhibitor I was used at 2 μM. All these inhibitors were purchased from Calbiochem (La Jolla, CA). To inhibit γ-secretase, γ-secretase inhibitor II (Calbiochem) was solubilized in DMSO (13 μM) added to PBS with Ca2+ and Mg2+.

shRNA constructs.

In designing MT1-MMP and MMP-2 shRNA, we used a PCR-based strategy for cloning hairpin sequences (38). A nonsilencing shRNA construct from Open Biosystems, whose sequence has been verified to contain no homology to known mammalian genes was used as a control. We used a single oligo which contains the hairpin and common 5′ (XhoI) and 3′ (EcoRI) ends as a PCR template. PCR was performed using universal primers containing the XhoI (5′ primer) and EcoRI (3′ primer) restriction sites. These PCR fragments were purified, and a restriction enzyme digestion with EcoRI and XhoI was performed. In addition, 2 μg of pSM2 vector (Open Biosystems, Huntsville, AL) were digested with EcoRI and XhoI restriction enzymes (New England Biolabs, Beverly, MA). The fragment was then gel purified using a 2% high-temperature-melting agarose (Sigma, St. Louis, MO). After the desired band was cut from the gel, it was purified using a QiaxII gel purification kit (Qiagen) and resuspended in 20 μl of 10 mM Tris. Using T4 DNA Ligase (New England Biolabs), 5 μl of cut PCR product and 1 μl of cut vector were ligated in a 20-μl ligation reaction. Ligations were transformed into PirPlus BW23474 chemically competent cells (Open Biosystems) using the manufacture's protocol. Recombinants were selected using both chloramphenicol (25 μg/ml) and kanamycin (25 μg/ml). Colonies were isolated and cultured, and the shRNA constructs were purified with an Endo-free maxi plasmid prep kit (Qiagen). shRNA constructs were then sequenced by using a U6 sequencing primer (GTAACTTGAAAGTATTTCG; Invitrogen) at the Texas A&M Gene Technologies Laboratory.

Transfection of shRNA into NRK cells.

At ∼60–80% confluence, shRNA constructs were transfected into NRK cells using the Arrest-In TM Transfection Reagent for RNAi (Open Biosystems). Briefly, each plate was transfected with 500 ng of the shRNA plasmid DNA and 20 ng of a nontargeted reporter and 20 ng of a luciferase reporter DNA in a total volume of 50 μl. For our experiments, luciferase activity in the cells was assessed to verify transfection. After 24–48 h of incubation, cells were assayed for reduction in protein expression by Western blot analysis. All controls were performed with the nonsilencing shRNA construct, whose sequence was verified to contain no homology to known mammalian genes.

Statistics.

Data are expressed as means ± SE. Groups were compared using ANOVA followed by post hoc t-tests with the Bonferroni correction. P < 0.05 was defined as significant.

RESULTS

Effects of proteinase inhibitors on ischemia-induced cadherin cleavage and loss.

Our previous results demonstrated that ischemia was associated with cleavage of E-cadherin to extracellular 80-kDa and intracellular 40-kDa fragments and loss of N-cadherin expression; however, the specific proteinase was not defined (7). Marambaud et al. (26) showed that γ-secretase cleaves E-cadherin into a similar-size fragment. Therefore, we utilized a γ-secretase inhibitor and examined ischemia-induced cadherin cleavage and loss. The γ-secretase inhibitor did not inhibit E-cadherin cleavage or the loss of N-cadherin, although the inhibitor itself was associated with a decrease in N-cadherin levels (Fig. 1). MMPs have also been shown to cleave cadherins (27, 36, 39). Therefore, to identify potential MMP candidates, we also examined chemical inhibitors of specific MMPs. MMP inhibitor I, which inhibits MMP-1, -3, -8, and -9, did not attenuate ischemia-induced disruption of E- and N-cadherin protein expression at 1.0 (data not shown), 30 (data not shown), or 150 μM (Fig. 1). The effects of MMP inhibitor III, which inhibits MMP-2 and -7, were evaulated next. At 10 nM, there was still a loss of full-length E-cadherin protein expression and a loss of N-cadherin protein expression, although the inhibitor partially preserved full-length E-cadherin (Fig. 1). To further examine whether MMP-2 plays a role in ischemic-induced disruption of cadherins, another inhibitor of MMP-2 was utilized. Blocking the activity of MMP-2 did not completely prevent loss of E- or N-cadherin (Fig. 1). Collectively, the data from MMP inhibitors suggest that MMP-1, -3, -8, or -9 do not play a role in ischemic disruption of cadherins, although a role for MMP-2 could not be excluded.

Fig. 1.

Impact of chemical inhibitors of proteinases on ishemia-induced disruption of cadherins. Before ischemia, cells were incubated for 30 min at desired concentration of inhibitors. MMP inhibitor I, an inhibitor of MMP-1 and -8 (1 μM), -9 (30 μM) and -3 (150 μM) was used at 150 mM; MMP inhibitor III, an inhibitor of MMP-1 (4 nM), MMP-2 (2.3 nM), MMP-13 (10 nM), MMP-7 (10 nM) and MMP-3 (135 nM), was used at 10 nM; and MMP-2 inhibitor I was used at 2 μM. All these inhibitors were purchased from Calbiochem (La Jolla, CA). To inhibit γ-secretase, γ-secretase inhibitor II (Calbiochem) was used at 13 μM. Ischemia was induced in NRK cells by layering cells with PBS supplemented with 1.5 mM CaCl2 and 2 mM MgCl2 followed by mineral oil. Lysates of ischemic NRK cells were prepared at 6 h, and Western blots were performed to assess E- and N-cadherin expression. The filled arrow marks full-length E-cadherin, while the open arrow identifies the intracellular 40-kDa fragment. Similar results were seen in 4 experiments.

Since TIMP-2 and -3, but not TIMP-1, inhibit MT1-MMP (10, 43, 54) and we previously showed that the loss of both E- and N-cadherin was inhibited by TIMP-3, but not TIMP-1 (7), the role of MT1-MMP in cadherin disruption was investigated. Initial experiments were designed to assess levels of active MT1-MMP following ischemia. Using an antibody that recognizes the activated form of MT1-MMP (9), an increase in active MT1-MMP was seen after 6 h of ischemia compared with control (Fig. 2). However, after 6 h of ischemia, there was a decrease in MMP-2 protein in the conditioned PBS (Fig. 2). Taken together, the inhibitor and protein expression experiments implicated MT1-MMP in cadherin cleavage during ischemia.

Fig. 2.

Impact of ischemia on MMP expression in NRK cells. Both conditioned PBS (MMP-2) and cell lysates (MT1-MMP) were isolated from control (Con) and ischemic (Isc) NRK cells, and Western blots were performed to assess MMP-2 and MT1-MMP expression. Similar results were seen in 4 experiments.

Regulation of MT1-MMP by the ECM.

Previous studies showed that the ECM regulates the activity of MT1-MMP (12, 45). Therefore, we examined the impact of the ECM on ischemia-induced cleavage and/or loss of E- and N-cadherin. A 40-kDa fragment of E-cadherin was seen in cells cultured on either collagen I or IV or laminin during ischemia (3–6 h), concomitant with an overall decrease in full-length E-cadherin (Fig. 3A). In contrast, the cells grown on fibronectin had no decrease in full-length E-cadherin expression and only a faint 40-kDa fragment after 3–6 h of ischemia (Fig. 3A). The expression of N-cadherin was also maintained only when NRK cells were cultured on fibronectin (Fig. 3A). Taken together, these data suggest that fibronectin is protective against ischemia-induced disruption of cadherins. Because NRK cells grown on fibronectin are protected from ischemia-induced disruption of cadherins, we examined levels of active MT1-MMP following ischemia in NRK cells cultured on the different matrices. After 6 h of ischemia, an increase in the levels of active MT1-MMP was seen in cells grown on laminin, collagen I, and collagen IV (Fig. 3B). In contrast, ischemia was not associated with increased active MT1-MMP in cells cultured on fibronectin, but rather a complete loss of active MT1-MMP expression (Fig. 3B). These results implicate MT1-MMP as a critical mediator of cadherin cleavage during ischemia.

Fig. 3.

Impact of the extracellular matrix on ischemia-induced cleavage of E-cadherin and loss of N-cadherin protein expression in NRK cells. A: NRK cells were grown on 4 different matrices: collagen I, collagen IV, laminin, and fibronectin. Ischemia was induced in NRK cells by layering cells with PBS supplemented with 1.5 mM CaCl2 and 2 mM MgCl2 followed by mineral oil. Lysates of ischemic NRK cells were prepared at the given times, and Western blots were performed to assess E- and N-cadherin expression. The filled arrow marks full-length E-cadherin, while the open arrow identifies the intracellular 40-kDa fragment. Similar results were seen in 4 replicate experiments. B: lysates of control and ischemic (6 h) NRK cells were prepared and Western blots were performed to assess MT1-MMP expression. Similar results were seen in 2 replicate experiments.

Expression of MT1-MMP in LE/LN cells.

L cells stably expressing E- or N-cadherin (LE and LN cells) were exposed to simulated ischemia for 6 h, and interestingly, in these cells no fragmentation and or/loss of either cadherin was seen (Fig. 4A). Because ischemia-induced disruption of cadherins did not occur, we examined MT1-MMP expression. Consistent with Figs. 1A and 3B, we observed an increased protein expression of active MT1-MMP after 6 h of ischemia in NRK cells; however, no expression of MT1-MMP was seen in control or ischemic LE or LN cells (Fig. 4B). These data further support the hypothesis that MT1-MMP is required for ischemia-induced disruption of cadherins.

Fig. 4.

Ischemia-induced cleavage of E- or N-cadherin in LE and LN cells. A: ischemia was induced in cells by layering cells with PBS supplemented with 1.5 mM CaCl2 and 2 mM MgCl2 followed by mineral oil. Lysates of ischemic LE and LN cells were prepared at 6 h, and Western blots were performed to assess E- and N-cadherin expression. Con, LE/LN are cells in normal growth media; PBS, cells incubated with PBS supplemented with Ca2+ and Mg2+ for 6 h; Isc, ischemic (6 h) cells. Similar results were seen in 4 replicate experiments. B: lysates of control (C) and ischemic (I) cells were prepared, and Western blots were performed to assess MT1-MMP expression. Similar results were seen in 2 replicate experiments.

Effects of MT1-MMP inhibition on cadherin expression and function.

To more directly examine whether MT1-MMP is involved in ischemia-induced disruption of cadherins, a blocking antibody and RNA interference strategy were used to block function and synthesis, respectively, of MT1-MMP. An antibody that recognizes the MT1-MMP catalytic domain and blocks function completely prevented ischemia-induced cleavage of E-cadherin, and loss of N-cadherin (Fig. 5).

Fig. 5.

Role of MT1-MMP in ischemia-induced cadherin cleavage in NRK cells. Before ischemia, cells were incubated for 30 min with the MT1-MMP blocking antibody (2.12 μg/ml). Lysates of ischemic NRK cells were prepared at 6 h, and Western blots were performed for E- and N-cadherin protein expression. The filled arrow marks full-length E-cadherin, while the open arrow identifies the intracellular 40-kDa fragment. Similar results were seen in 2 replicate experiments.

To suppress MT1-MMP and MMP-2 protein expression, shRNA constructs were developed. Three target cassettes were synthesized per target gene (Fig. 6A) and inserted into the pSM2 vector. Once the vectors were sequenced, they were transfected into NRK cells. The negative control contains the pSM2 vector with a sequence that has been verified to contain no homology to mammalian genes. Using Western blot analysis, protein expression of MT1-MMP and MMP-2 was examined in transfected NRK cells at 24 and 48 h. The second template (Temp 2) of the PCR cassettes to target MMP-2 ablated MMP-2 protein expression after 48 h in NRK cells (Fig. 6B). In contrast, Temp 1 and Temp 3 did not change MMP-2 protein expression. Similarly, after 48 h, the Temp 2 of MT1-MMP PCR cassettes inhibited MT1-MMP protein expression (Fig. 6B). Although Temp 1 decreased MT1-MMP protein expression, it was not as efficient as Temp 2 and there was no apparent change in MT1-MMP protein levels with Temp 3. Importantly, the MMP-2 and MT1-MMP shRNA constructs were specific for each gene; i.e., MT1-MMP shRNA did not affect MMP-2 levels and MMP-2 shRNA did not affect MT1-MMP protein expression (Fig. 6C). Therefore, for subsequent shRNA experiments, the Temp 2 PCR cassettes were used to ablate MMP-2 and MT1-MMP protein expression.

Fig. 6.

MMP-2 and MT1-MMP protein expression of NRK cells transfected with shRNA constructs. A: sequences of the 3 targets for MMP-2 and MT1-MMP. B: NRK cells were transfected with 3 different templates of PCR cassettes that target MMP-2 (M2) or MT1-MMP (MT1). Conditioned PBS and NRK lysates were isolated at the indicated times, and Western blots were performed to assess MMP-2 and MT1-MMP expression. Similar results were seen in 2 replicate experiments. C: NRK cells were transfected with shRNA (Temp 2) against MMP-2 (M2) or MT1-MMP (MT1). Forty-eight hours later, conditioned PBS and NRK lysates were isolated and Western blots were performed to assess MMP-2 and MT1-MMP expression. Similar results were seen in 2 replicate experiments.

Next, cadherin and catenin protein expression was examined in ischemic NRK cells transfected with shRNA constructs. Inhibiting MT1-MMP protein expression with shRNA resulted in attenuation of ischemia-induced cleavage and loss of protein expression (Fig. 7). The transfection of shRNAs, the control (pSM2), MMP-2 (pSM2-M2), and MT1-MMP (pSM2-MT1), did not effect the protein expression of cadherins and catenins under control conditions, suggesting that MMP-2 and MT1-MMP were not required for cadherin/catenin expression (Fig. 7). Ischemia-induced loss of E-cadherin was seen in cells transfected with the pMS2 (−) and pSM2-M2 (M2) vector at 6 h but did not occur in NRK cells with the pSM2-MT1-MMP (MT1) vector (Fig. 7). Although we did observe the intracellular 40-kDa fragment in these experiments, the intensity of the band was less than seen in other studies. In addition, there was neither ischemia-induced loss of N-cadherin nor loss of α-catenin as seen in control and MMP-2 shRNA cells. Similar to previous data, there was no change in β-catenin or p120 protein expression in all experimental groups (7). In the absence of MT1-MMP protein expression, ischemic disruption of cadherins/catenins is not seen, suggesting that MT1-MMP is required for ischemia-induced disruption of the cadherin/catenin complex.

Fig. 7.

Cadherin/catenin protein expression in ischemic NRK cells transfected with shRNAs. Two days before ischemic insult, NRK cells were transfected with a control (−), MMP-2 (M2), or MT1-MMP (MT1) shRNA vector. Lysates of control and ischemic NRK cells (6 h) were prepared, and Western blots were performed to assess cadherin and catenin expression. The filled arrow marks full-length E-cadherin, while the open arrow identifies the intracellular 40-kDa fragment. Similar results were seen in 4 replicate experiments.

Cadherin/catenin localization in pSM2 and pSM2–14 NRK cells was evaluated by immunofluorescence microscopy. While normal pSM2 NRK cells expressed E-cadherin at the plasma membrane, 6 h of ischemia resulted in a decrease in E-cadherin at the cell membrane (Fig. 8). In pSM2-MT1 cells, E-cadherin expression remained at the plasma membrane after 6 h of ischemia (Fig. 8). N-cadherin expression in normal pSM2 cells is localized to the cell membrane, but after 6 h of ischemia N-cadherin staining was lost. Again, pSM2-MT1 maintained cell membrane localized N-cadherin staining during ischemia. After 6 h of ischemia, there was a loss of α-catenin, but not β- and p120-catenin, labeling in pSM2 cells (Fig. 8). In pSM2-MT1 cells, α-catenin is localized at the plasma membrane under both control and ischemic conditions (Fig. 8). There was no difference in cadherin and catenin localization when MT1-MMP expression was inhibited, suggesting that MT1-MMP is not required for basal cadherin expression or localization.

Fig. 8.

Impact of ischemia on cadherin/catenin localization in pSM2 and pSM2-MT1 NRK cells. Ischemia was induced in NRK transfected with control (pSM2) or MT1-MMP (pSM2-MT1) shRNA constructs. Cells were fixed in 2% paraformaldehyde, 1° antibody (1:100) incubated overnight at 4°C, 2° antibody (1:200) incubated 1 h at room temperature, and detection via immunofluorescence. Prolong antifade with DAPI was used to counterstain the nuclei. The width of each field is 240 μm. Similar results were seen in 2 replicate experiments.

To examine the role that MT1-MMP plays in cadherin- mediated cell-to-cell adhesion, cell morphology and cell aggregation were assessed. pSM2-transfected NRK cells viewed by phase-contrast microscopy revealed closely packed, polygonally shaped cells with little light transmitted between them (Fig. 9A). Following 6 h of ischemia, pSM2 cells exhibited a loss of cell-to-cell adhesion and increased transmission of light at cell-to-cell boundaries, although cells remained attached to the growing surface (Fig. 9A). pSM2-MT1-transfected NRK cells were similar to controls and even after 6 h of ischemia the cell morphology was not changed (Fig. 9A). Using the classic aggregation assay developed for cadherins (11), cadherin function in ischemic NRK cells transfected with pSM2 and pSM2-MT1 was examined. Control and ischemic NRK cells that were transfected with either pSM2 or pSM2-MT1 were allowed to aggregate for 20 and 60 min at 37°C while shaking at 80 rpm. A significant decrease in cell aggregation induced by 6 h of ischemia was seen at both 20 (27 ± 3% of control) and 60 min (47 ± 4% of control) of cell aggregation in pSM2 cells (Fig. 9B). However, pSM2-MT1 cells did not exhibit a reduction in cell aggregation after 6 h of ischemia (Fig. 9B). Collectively, these data suggest that MT1-MMP plays an essential role in ischemia-induced disruption of cell aggregation and cell-to-cell contact.

Fig. 9.

Impact of ischemia on cell-to-cell adhesion in NRK cells transfected with pSM2 and pSM2-MT1. NRK cells were transfected with pSM2 or pSM2-MT1 48 h before ischemia. A: morphology of control and 6-h ischemic NRK cells tranfected with pSM2 or pSM2-MT1 were assessed by phase-contrast microscopy. The width of each field is 1,600 and 400 μm. Similar results were seen in 2 replicate experiments. B: after ischemia, cells were dissociated as described in materials and methods and allowed to aggregate for 20 and 60 min. Number of cell aggregates was converted to %control. *Significant difference (P < 0.05) from positive control cell aggregation values at the corresponding time (n = 10 for each condition).

DISCUSSION

Because ischemia is a leading cause of ARF, investigating the targets of ischemic insult will provide insight into the mechanisms underlying ARF. The cadherin/catenin complex, which is critical to renal proximal epithelial cell function, is disrupted by ischemic insult. In this study, an ischemia model which resembles ischemic renal injury in vivo resulted in selective fragmentation/loss of E-cadherin and loss of N-cadherin expression that could be blocked by inhibition of MT1-MMP. These data suggest the MT1-MMP represents a novel therapeutic target in the treatment of ARF.

Among the MMPs so far known, selective inhibition by TIMP-2 or -3 but not by TIMP-1 was reported only for MT1-MMP and MT3-MMP (10, 43, 54). An antibody that recognizes the catalytic domain and blocks function of MT1-MMP completely prevented ischemia-induced cleavage of E-cadherin and loss of N-cadherin. In addition, an shRNA construct that targets MT1-MMP blocked ischemia-induced cadherin disruption. In contrast, MMP-2 shRNA did not suppress ischemia-induced cleavage and/or loss of E- and N-cadherin, suggesting that MMP-2 does not play a role in cadherin disruption. These results indicate that disruption of the complex during ischemia occurs directly via MT1-MMP, rather than by MT1-MMP activation of MMP-2. Alternatively, MT1-MMP could be activating another proteinase which cleaves the cadherins such as ADAM10 which has recently been shown to cleave N-cadherin (41).

Integrins and ECM substrates have been shown to influence MT1-MMP activation (8). Unlike cells grown on laminin, collagen I, and collagen IV, there was not an increase in expression of active MT1-MMP under ischemic conditions in NRK cells grown on fibronectin. Our results are consistent with previous findings showing that fibronectin inhibited expression of active MT1-MMP in human breast cancer cells (45) and endothelial cells (12). Furthermore, ischemia-induced disruption of cadherins did not occur in LE or LN cells which were deficient in basal and ischemia-inducible MT1-MMP, further supporting the role of MT1-MMP in ischemia-induced cadherin disruption.

The role of membrane-bound metalloprotease in the cleavage of E-cadherin was first suspected by Ito et al. (15) using a conditioned media approach in cancer cell lines that had relatively high and low E-cadherin cleavage. Shedding of E-cadherin in apoptotic MDCK cells also could not be prevented by an inhibitor of soluble MMPs (43). It was proposed that this cleavage could be due to a MMP with the same substrate specificity as that reported by Ito et al. (15). A similar pattern of inhibition was reported for the extracellular cleavage of VE-cadherin during apoptosis (14). More direct evidence for MT1-MMP in cadherin cleavage comes from Rozanov et al. (42). In these studies, abundance of an 84-kDa E-cadherin fragment in the culture medium of MCF-7 cells was attenuated by the expression of MT1-MMP mutants. In addition, in human lung tumor cells a soluble 80-kDa fragment of E-cadherin induced MT1-MMP expression that was mostly present in the active form (35). Full-length E-cadherin expression has also been shown to downregulate MT1-MMP expression in tumor cells (1, 20, 35). Although we detected a similar 80-kDa fragment, our data suggest that MT1-MMP cleaves E-cadherin to release an 80-kDa extracellular fragment, which could then increase active MT1-MMP in ischemic NRK cells.

The importance of cadherin cleavage during ischemia-reperfusion injury was first shown by Bush et al. (4); however, these studies used a severe (3 h) ischemic insult in vivo. Recent data indicate that MMPs play a role in ischemic ARF, as MMPs were increased in a rat postischemic kidney tissue and were localized to the renal tubules (2). In a more recent report, it was shown that 24 h after reperfusion following 30 min of ischemia, full-length N-cadherin expression was absent in mouse kidneys; however, N-cadherin fragments were seen in kidney lystates (30). In our studies, we did not detect N-cadherin fragments despite using a number of commerically available antibodies; however, we are able to preserve full-length N-cadherin with MMP inhibitors, as well as with the MT1-MMP blocking antibody or shRNA. Our results are similar to those of Pon et al. (40), who demonstrated that inhibition of MMP-2 decreased N-cadherin turnover, but were unable to detect fragments. With regard to MT1-MMP and ischemia, it has been shown that MT1-MMP activity increases during ischemia and reperfusion in the myocardium in vivo (9). In addition, increased active MT1-MMP was also observed in acute hindlimb ischemia (31).

In summary, the present study demonstrates a role of MT1-MMP in ischemia-induced disruption of cadherin/catenin complexes. Ischemia stimulated expression of active MT1-MMP. Under conditions of cell attachment to fibronectin that attenuated cadherin loss during ischemia, levels of active MT1-MMP were not increased by ischemia; and in cells (LE and LN) that did not display cadherin loss during ischemia, MT1-MMP was not expressed. Importantly, ischemia-induced cadherin disruption is attenuated by interfering with MT1-MMP expression (shRNA) or function (blocking antibody). Furthermore, knockdown of MT1-MMP with shRNA constructs restored cadherin function and cell-cell adhesion in ischemic NRK cells. Collectively, these data suggest that MT1-MMP is required for ischemia-induced disruption of cadherins and catenins.

GRANTS

These studies were supported by the Department of Medical Pharmacology and Toxicology and the Center for Environmental and Rural Health (P30-ES09106).

Acknowledgments

The authors thank Dr. J. Balsamo (University of Iowa) for providing the LE cells.

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

REFERENCES

View Abstract